In the field of image-forming devices, it has been customary to perform a calibration process in order to maintain image quality. The calibration process is executed in order to prevent changes in the density of toner images that occur through extended use of the device or due to environmental changes. One such calibration process disclosed in Japanese unexamined patent application publication No. HEI-4-77060A entails forming density patches (test patterns), measuring the densities of the density patches, and calibrating image densities in image formation based on these measurements.
This calibration process includes the steps of measuring a plurality of density patches having different densities to produce measured output values, and forming ratio data (calibration data) for offsetting a difference between these measured output values and reference output values. Since numerous values for ratio data are required for each density level within the overall density range, the problem becomes how to acquire so many values of ratio data. In the example of Japanese unexamined patent application publication No. HEI-4-77060A, the image-forming device forms test patterns for all density levels and measures the densities of all these levels. The image-forming device then acquires calibration coefficients for calibrating the input values (measured values) and output values. However, a method of measuring the densities of numerous density levels can slow the process and consumes a lot of developer or other consumables.